Assessment of Candidate Molten Salt Coolants for the Advanced High Temperature Reactor (AHTR)

Abstract

The Advanced High-Temperature Reactor (AHTR) is a novel reactor design that utilizes the graphite-matrix high-temperature fuel of helium-cooled reactors, but provides cooling with a high-temperature fluoride salt. For applications at temperatures greater than 900 C the AHTR is also referred to as a Liquid-Salt-Cooled Very High-Temperature Reactor (LS-VHTR). This report provides an assessment of candidate salts proposed as the primary coolant for the AHTR based upon a review of physical properties, nuclear properties, and chemical factors. The physical properties most relevant for coolant service were reviewed. Key chemical factors that influence material compatibility were also analyzed for the purpose of screening salt candidates. Some simple screening factors related to the nuclear properties of salts were also developed. The moderating ratio and neutron-absorption cross-section were compiled for each salt. The short-lived activation products, long-lived transmutation activity, and reactivity coefficients associated with various salt candidates were estimated using a computational model. Table A presents a summary of the properties of the candidate coolant salts. Certain factors in this table, such as melting point, vapor pressure, and nuclear properties, can be viewed as stand-alone parameters for screening candidates. Heat-transfer properties are considered as a group in Sect. 3 in order to evaluate the combined effects of various factors. In the course of this review, it became apparent that the state of the properties database was strong in some areas and weak in others. A qualitative map of the state of the database and predictive capabilities is given in Table B. It is apparent that the property of thermal conductivity has the greatest uncertainty and is the most difficult to measure. The database, with respect to heat capacity, can be improved with modern instruments and modest effort. In general, ''lighter'' (low-Z) salts tend to exhibit better heat transfer and nuclear performance metrics. Lighter salts also tend to have more favorable (larger) moderating ratios, and thus should have a more favorable coolant-voiding behavior in-core. Heavy (high-Z) salts tend to have lower heat capacities and thermal conductivities and more significant activation and transmutation products. However, all of the salts are relatively good heat-transfer agents. A detailed discussion of each property and the combination of properties that served as a heat-transfer metric is presented in the body of this report. In addition to neutronic metrics, such as moderating ratio and neutron absorption, the activation properties of the salts were investigated (Table C). Again, lighter salts tend to have more favorable activation properties compared to salts with high atomic-number constituents. A simple model for estimating the reactivity coefficients associated with a reduction of salt content in the core (voiding or thermal expansion) was also developed, and the primary parameters were investigated. It appears that reasonable design flexibility exists to select a safe combination of fuel-element design and salt coolant for most of the candidate salts. Materials compatibility is an overriding consideration for high-temperature reactors; therefore the question was posed whether any one of the candidate salts was inherently, or significantly, more corrosive than another. This is a very complex subject, and it was not possible to exclude any fluoride salts based on the corrosion database. The corrosion database clearly indicates superior container alloys, but the effect of salt identity is masked by many factors which are likely more important (impurities, redox condition) in the testing evidence than salt identity. Despite this uncertainty, some reasonable preferences can be recommended, and these are indicated in the conclusions. The reasoning to support these conclusions is established in the body of this report.